Download Integrating mechanistic and evolutionary analysis of life history

OUP CORRECTED PROOF – FINAL, 04/23/2011, SPi
CHAPTER 1
Integrating mechanistic and
evolutionary analysis of life history
variation
Christian Braendle, Andreas Heyland, and Thomas Flatt
1.1
Introduction
Life histories—describing essential patterns of
organismal growth, maturation, reproduction, and
survival—show tremendous variation across individuals, populations, species, and environments.
Understanding this variation is the goal of life history research. The analytical framework of life history
theory focuses on the variation and interaction of
different key maturational, reproductive, and other
demographic traits, given that natural selection acts
to maximize fitness of a life history as a whole (Roff
1992, Stearns 1992). Fitness integrates over the entire
reproductive performance of the organism, and life
history traits are the major fitness components
underlying this integration. However, the investment into alternative life history traits, and thus the
possible set of trait combinations, is restricted by
genetic, developmental, physiological, and phylogenetic limits. Apart from explaining variation in
life history strategies as a result of natural selection,
identifying how such trade-offs and constraints
shape life histories is the central aim of life history
research.
In this chapter we introduce the basic concepts
and definitions of life history theory and argue for
the importance of integrating a mechanistic perspective into research on life histories. While most
traditional life history research is based on mathematical, statistical, and phylogenetic approaches
without explicit reference to underlying mechanisms, today’s principal research challenge is to
fill this gap through experimental characteriza-
tion of the proximate basis of life histories. The
analysis of genetic, developmental, and physiological factors that shape life history traits will
ultimately allow us to determine how evolutionary changes in such mechanisms generate, facilitate, or constrain the diversification of life histories.
Integrating mechanistic and evolutionary analyses of life history variation is part of a global quest
in biology that seeks a shared understanding
of proximate and ultimate causes of phenotypic
variation.
1.2 The life history framework
1.2.1 What is a life history?
A life history encompasses the life of an individual
from its birth to its death, describing the age- or
stage-specific patterns of maturation, reproduction,
survival, and death. The major objective of life history research is to understand how evolution, given
selection imposed by ecological challenges, shapes
organisms to achieve reproductive success. The second objective of life history research is to understand whether and how, given internal trade-offs
and constraints, selection can optimize a set of life
history traits to maximize reproductive success.
Since organisms dispose of limited resources, which
must be competitively allocated to differing functions, such as growth, reproduction, survival, and
maintenance, resources invested into one function
cannot be invested into another, leading to tradeoffs. In addition, life history research explores
3
OUP CORRECTED PROOF – FINAL, 04/23/2011, SPi
4
M E C H A N I S M S O F L I F E H I S T O RY E V O L U T I O N
taxon-specific features of life cycles and life history
decisions, including patterns of sex allocation, alternative phenotypes, or larva-to-adult transitions. For
in-depth treatments of the evolution of life histories
and life history theory see Stearns (1992), Roff (1992,
2002), and Charlesworth (1994).
1.2.2
Life history traits and fitness
Life history traits represent quantitative, demographic properties of organisms that are directly
related to the two major components of fitness, i.e.,
survival and reproduction. Classical life history
analysis considers the following to be the principal
life history traits (Stearns 1992):
•
•
•
•
•
•
•
size at birth
growth pattern
age and size at maturity
number, size, and sex ratio of offspring
age-and size-specific reproductive investments
age- and size-specific mortality schedules
length of life.
These traits essentially represent the demographic
parameters required to estimate fitness as defined by
the Malthusian parameter (or similar fitness measures). The Malthusian parameter (also called the
instantaneous rate of natural increase, r) is the solution to the Euler–Lotka equation, which describes
population growth by summing reproductive events
and survival probabilities over the entire lifetime of
individuals (Stearns 1992). Thus, life history traits are
directly linked to fitness, with fitness being defined
by population growth models from demography.
In contrast to classical life history traits, morphological, physiological, or behavioral traits are considered to contribute to fitness only indirectly (e.g.,
Roff 2007b). However, this distinction is somewhat
arbitrary. For example, certain morphological traits
such as body size or gonad size may correspond to
life history traits (or at least are correlates thereof).
In the literature, the term “life history trait” is often
used interchangeably with fitness components, so
that many phenotypic characters with major effects
on reproduction and survival have been called life
history traits.
Because of their complexity and demographic
nature, life history traits are usually treated as quan-
titative, polygenic traits (Falconer and MacKay
1996). The expression of life history traits is also
highly contingent on the environment, so that life
history research places particular emphasis onto the
concept of phenotypic plasticity, i.e., the ability of a
single genotype to produce different phenotypes
across environments (Stearns 1992). Plasticity is
described by “reaction norms”, mathematical functions that relate the phenotypic values adopted by a
given genotype to changes in the environment.
Selection shapes life history plasticity by acting on
genetic variation for plasticity, which is present
when the reaction norms that represent different
genotypes are non-parallel across the same range of
environments (so-called genotype by environment
interactions, or G × E). Reaction norms (and thus
plasticity) are considered to be optimal when they
maximize fitness for each of the different environments (Stearns and Koella 1986).
1.2.3 Trade-offs and constraints
A key postulate of life history theory is that the values and combinations of life history traits are limited by factors internal to the organism, namely
trade-offs and constraints. These intrinsic factors
ultimately limit and direct the evolutionary response
to the external force of selection. A life history tradeoff occurs when an increased investment in one fitness component causes a reduced investment in
another fitness component, i.e., a fitness benefit in
one trait exacts a fitness cost in another. Examples of
classical life history trade-offs are survival versus
reproduction, number versus size of offspring, or
current reproduction versus future reproduction
(Stearns 1992).
Trade-offs are usually described as phenotypic or
genetic covariances or correlations among traits,
without reference to their causal relationships. If the
relationship can be shown to be genetic, negative
genetic covariance among traits is expected to limit
the evolution of each of these traits. Such genetic or
evolutionary trade-offs are considered at the population level, i.e., as defined by genetic correlations
among individuals or correlated phenotypic
responses to selection. Genetic trade-offs are traditionally assumed to stem from antagonistic pleiotropy or linkage disequilibrium. These trade-offs
OUP CORRECTED PROOF – FINAL, 04/23/2011, SPi
I N T E G R A T I N G M E C H A N I S T I C A N D E V O L U T I O N A RY A N A LY S I S O F L I F E H I S T O RY V A R I A T I O N
also manifest themselves at the physiological or
individual level, for example when an individual
with increased reproductive effort in one year
exhibits a reduction in reproductive output in the
next year. Such physiological trade-offs are thought
to be due energy limitations, i.e., the allocation of
resources among competing functions. Importantly,
trade-offs may exist at population level, but not at
individual, physiological level (Stearns 1989, Houle
1991, Stearns 1992).
In contrast to trade-offs, the term “constraint” is
often used to described absolute limits to or biases
upon trait expression and combination. Constraints
may describe physical factors, developmental properties, or historical contingencies that prevent an
organism from expressing a certain phenotype or a
population from attaining a certain fitness optimum
in response to selection (Maynard Smith et al. 1985).
The distinction between trade-offs and constraints
is not strict, and trade-offs are often regarded as one
type of constraint. In the life history context, constraints usually refer to phylogenetic, lineage-specific characteristics that impose absolute limits on
trait expression in a given organismal group.
1.2.4 Empirical approaches in life history
research
Although classic life history analysis has been
largely theory-driven, much empirical research has
addressed the questions and predictions raised by
life history theory, using both non-genetic and
genetic approaches (Stearns 1992, Roff 1992, 2002,
2007b; also see Chapter 2). Non-genetic approaches
include phenotypic correlations to examine patterns
of life history trait covariation among populations
and species, experimental phenotypic manipulations,
and statistical tools from comparative analysis to
control for phylogenetic history. Genetic approaches
to the study of life history variation are predominantly based on the framework of quantitative
genetics. Most of this work has concentrated on the
detection and analysis of genetic trade-offs, either
through the study of covariances and correlations
among life history traits between relatives (e.g.,
pedigree analyses) or through correlated responses
of life history traits to artificial selection or experimental evolution. This research framework has
5
generated a substantial body of empirical evidence
that has revealed how selection operates on life history traits, contingent on the environment and
trade-offs (Stearns 1992, Roff 1992, 2002, 2007a,b).
Despite these extensive efforts, very few studies
have examined the mechanistic underpinnings of
life history traits. For example, inferred interrelationships among life history traits rarely describe
more than statistically determined associations.
A major limitation common to the classical
approaches in life history research is therefore the
ignorance of the proximate causes that determine or
modulate life histories and their evolution.
1.3 The study of causal mechanisms
linking genotype to phenotype
Understanding how a genotype translates into a
phenotype is one of the most fundamental problems in biology. In most cases, phenotypes cannot
be simply inferred from their underlying genotypes,
and vice versa, because the mapping of genotypes
onto phenotypes is often a non-linear process,
shaped by a multitude of complex genetic and environmental interactions. Moreover, a single genotype
may generate multiple phenotypes and, conversely,
multiple genotypes may generate a single phenotype.
That such properties of the genotype–phenotype
map are relevant for our understanding of the evolutionary process has been emphasized for a long
time (e.g., Lewontin 1974, Houle 2001), but it is only
relatively recently that the causal relationships
between genotype and phenotype have received
increased attention from evolutionary biologists
(e.g., Pigliucci 2010). While research at the interface
of development and evolution has begun to tackle
the significance of the genotype–phenotype map in
morphological evolution, the causal connection
between genotypes and phenotypes for fitness components is still extremely rudimentary (e.g., Chapter 2
and Roff 2007b).
Traditionally, attempts to link the genotype with
the phenotype have been regarded as the principal
task of “reductionist” branches of biology, including molecular, cellular, and developmental biology.
Developmental genetics in particular has emerged
as the prime discipline in connecting gene function
during development with phenotypic outcomes,
OUP CORRECTED PROOF – FINAL, 04/23/2011, SPi
6
M E C H A N I S M S O F L I F E H I S T O RY E V O L U T I O N
primarily by relying on mutational analysis and
forward genetics. The great power of this approach
lies in the typically high degree of causal inference
that can be made through carefully controlled
manipulation of isolated genetic factors and their
phenotypic effects. The general downside of this
approach is that such studies are generally limited
to the study of single, highly pleiotropic mutations
with large phenotypic effects. In addition, developmental genetic analyses are generally limited to the
study of a single or or a small number of laboratory
populations in highly simplified artificial environments, aiming to reduce variation engendered by
genetic background or environmental context as
much as possible. This research approach starkly
contrasts with that of evolutionary biologists, whose
primary concern is the study of quantitative genotypic and phenotypic variation among populations
or species. Here, in contrast to developmental genetics, the inferred genotype–phenotype relationships
are generally of indirect, associative nature, rarely
permitting inferences about the causal connections
between genotypic and phenotypic variation.
As advocated in many chapters throughout this
book, a better future understanding of many issues
in life history evolution will require the integration
of evolutionary and organismal biology with molecular and developmental biology (e.g., Dean and
Thornton 2007). That unfortunate historical separations between biological disciplines can be overcome
is well illustrated by the successful rapprochement
of evolutionary and developmental biology (e.g.,
Raff and Kaufman 1983, Carroll et al. 2000, Stern
2010). Although initially mainly concerned with the
description of evolutionary diversification or conservation of developmental mechanisms, the central
aim of evolutionary developmental biology (evodevo) has recently shifted to the experimental analysis of how properties of genetic and developmental
architecture impact phenotypic evolution. Evo-devo
therefore addresses specific issues directly relevant
to the understanding of life history evolution, such
as the mechanistic basis of developmental biases
and constraints or phenotypic plasticity. More generally, as life history traits are high-level phenotypes
that depend on the ensemble of morphological and
physiological traits, the mechanistic analysis of life
history evolution can consequently be regarded as
an extension of the principal objective of evo-devo,
namely to understand which developmental and
genetic changes underlie phenotypic evolution.
Uncovering the mechanistic basis of life history
variation is a non-trivial challenge. Life history
traits were defined by evolutionary ecologists with
the intent of reducing phenotypic complexity by
focusing on a small number of traits that summarize the essential fitness components and by ignoring the underlying genetic, developmental, and
physiological mechanisms that govern the expression of these traits. A given life history trait can thus
be thought of as a functionally complex phenotype
resulting from the integration of a suite of morphological, physiological, or behavioral phenotypes. At
the level of the individual, their characteristics have
therefore to be understood in terms of both the construction of multiple individual traits as well as
their spatial and temporal integration into a higherlevel phenotype. As such, life history traits are a priori composite, quantitative, polygenic traits whose
expression is often highly contingent upon plasticity, pleiotropy, and epistasis. All these properties
render the mechanistic analysis of life history traits
extremely difficult in practice.
1.4 How can mechanistic insights
contribute to understanding life history
evolution?
Despite the inherent difficulties in studying the
proximate basis of life histories, considerable
progress has been made in our mechanistic understanding of life history evolution, with major contributions stemming from molecular genetic studies
on experimental model organisms. Here we briefly
discuss the importance of integrating such mechanistic information into organismal life history
research; many more detailed examples can be
found throughout the chapters in this book. For further reading on integrative approaches in life history biology we recommend the reviews by Houle
(2001), Leroi (2001), Barnes and Partridge (2003),
Harshman and Zera (2007), Chapter 5 in Van
Straalen and Roelofs (2006), Roff (2007b), and Flatt
and Schmidt (2009).
OUP CORRECTED PROOF – FINAL, 04/23/2011, SPi
I N T E G R A T I N G M E C H A N I S T I C A N D E V O L U T I O N A RY A N A LY S I S O F L I F E H I S T O RY V A R I A T I O N
1.4.1 Why understanding mechanisms is
important for answering evolutionary questions
While it is clear that knowledge of the proximate
basis of life histories does not provide information
about the ecological or evolutionary relevance of
such mechanisms, it enables evolutionary biologists
to address several fundamental questions about life
history evolution, including, for example:
• What is the function of genes that are genetically
variable in natural populations and that contribute
to ecological adaptation?
• Are major candidate genes, as identified by
molecular genetics, variable in natural populations?
• If so, do polymorphisms at these loci actually
contribute to the evolution of life history traits in
the wild?
• Are the genes that impact life history evolutionarily conserved or lineage-specific?
• What genetic and physiological mechanisms
determine or modulate the expression of ecologically and evolutionarily important trade-offs?
• Are such trade-offs, as commonly assumed,
resource based, or are they due to mechanisms
independent of energy allocation?
• What are the mechanisms that mediate life
history plasticity?
1.4.2 The molecular identity and function
of genes that affect life history
Studies in molecular and developmental genetics
inform us about the molecular identity and function
of genes, including those that affect life history traits
and other fitness components. The functionally bestunderstood genes that affect life history traits have
been analyzed in model organisms such as
Arabidopsis, Drosophila, or C. elegans. Information
about the function of such genes is useful, for example, when evolutionary biologists want to investigate the consequences of allelic variation at such loci
in natural populations. Although natural alleles
might have much more subtle phenotypes than laboratory induced mutant alleles, detailed knowledge
about gene function might help organismal biologists to understand whether and how particular
7
genes contribute to ecologically relevant phenotypes
and thus why selection acts on such loci. This does
not mean that every gene with a major phenotypic
effect on a fitness-related trait, as identified by
molecular genetics, is in fact ecologically or evolutionarily relevant in natural populations; many such
genes might not harbor standing genetic variation
affecting life history phenotypes and might therefore not contribute to evolutionary change in the
wild. Yet, it is also clear that loci that do contribute to
phenotypic variation in fitness-related traits and
thus to ecological adaptation in natural populations
are a subset of all genes, including those that have
been functionally studied by molecular geneticists
(e.g., Stern 2000, Flatt 2004, Flatt and Schmidt 2009).
While developmental and molecular genetic
approaches do inform us about the ecological or
evolutionary significance of specific genes, they
have proved powerful in identifying the molecular
mechanisms that affect life history traits, for instance
their endocrine regulation (Tatar et al. 2003,
Fielenbach and Antebi 2008). Perhaps the best
examples are genes known to affect adult survival
and longevity in the nematode, fruit fly, and mouse;
these have received particular attention, not only
from biomedical researchers because of their potential implications for human gerontology (see
Chapter 16), but also from evolutionary biologists
because of their potential relevance for understanding the evolution of aging. During the past 20 years,
numerous mutations that extend lifespan have been
identified in diverse model organisms (e.g., Kenyon
2010; also see Chapter 14). Many of these mutations
were found to affect a key metabolic pathway—the
insulin/insulin-like growth factor signaling pathway—indicating that decreased effectiveness of
insulin/IGF-like signaling causes lifespan extension, linked to correlated responses in reproduction,
growth, and metabolism. These pivotal discoveries,
many of which are discussed in this book, not only
demonstrate the feasibility of molecular genetic
analyses of complex life history traits such as
lifespan, but also suggest that certain evolutionarily
conserved signaling pathways are potential key
regulators of major life history traits (also see
Chapters 27 and 28). Many of these findings have
also contributed to our understanding of life history
OUP CORRECTED PROOF – FINAL, 04/23/2011, SPi
8
M E C H A N I S M S O F L I F E H I S T O RY E V O L U T I O N
trade-offs (see below and Chapters 11 and 13). The
molecular genetic analysis of lifespan has thus rapidly become of great interest to many researchers
studying life histories, and this interest is now paving the way for an integration of mechanistic and
evolutionary approaches towards the understanding of life history variation (e.g., Partridge and
Gems 2006, Flatt and Schmidt 2009).
In addition to functional studies of individual
mutations, genome-wide gene expression analyses
have also been widely used by both molecular and
evolutionary biologists to investigate the proximate
basis of life history variation (as is discussed in
detail in Chapter 2). For example, genome-wide
transcriptional profiling has been used to identify
candidate genes involved in lifespan regulation
(e.g. Murphy et al. 2003), or to describe gene expression patterns associated with particular life history
stages, for example dauer larva formation in C. elegans (Wang and Kim 2003). Many of these studies
illustrate the complex and manifold changes in gene
expression associated with life history variation and
further indicate that life history trade-offs might
emerge through “conflicts over gene expression”,
i.e., antagonistic pleiotropic effects of genes involved
in multiple functions (Stearns and Magwene 2003,
Bochdanovits and de Jong 2004). However, the
functional interpretation of such data remains challenging because the precise causal connections
between transcriptional changes and the resulting
phenotypes are rarely known. Thus, while it is clear
from these few examples that we have learned a
great deal about the molecular genetic basis of life
history traits, a current key challenge is to integrate
such mechanistic insights into the evolutionary
framework (also see Chapters 27 and 28). One obvious question for the evolutionary biologist is, for
example, whether the candidate genes identified by
molecular geneticists actually matter in natural
populations.
1.4.3 Are candidate life history genes
ecologically and evolutionarily relevant?
Mutational, transgenic, and genomic analyses in
model organisms have been successful in identifying at least some of the key mechanisms that affect
life history traits. However, while many of these
mechanisms show a surprisingly high degree of
conservation across widely divergent taxa, their relevance in shaping evolutionary life history variation in natural populations is not yet sufficiently
clear. Determining whether and how such mechanisms evolve to generate natural life history variation represents a promising starting point for the
integration of functional and evolutionary analysis
of life histories. In most cases, however, such studies are limited to model organisms. Such an analysis
requires testing of whether the genes involved in
these candidate mechanisms show actual variation
in natural populations and, as a more challenging
step, to functionally demonstrate that this allelic
variation impacts the life history trait in question.
Several studies suggest that genes identified
through molecular and developmental genetic
analyses indeed harbor natural allelic variation that
contributes to population variation in life history
traits, for example in Drosophila (e.g., Schmidt et al.
2000, Paaby and Schmidt 2008, Paaby et al. 2010;
also see Chapter 18), or in Arabidopsis (e.g., Todesco
et al. 2010; also see Chapter 9). Although the screening of natural polymorphisms in candidate life history genes only provides a first glimpse of the
molecular basis of life history variation, such initial
findings are encouraging since they indicate that
developmental and molecular genetic studies
indeed generate valuable candidate genes of interest for evolutionary biologists.
In contrast to the analysis of natural allelic variants at major candidate loci identified by molecular
and developmental genetics, quantitative trait locus
(QTL) mapping provides a less biased, yet technically challenging, approach to the characterization
of the genetic basis of polygenic quantitative traits,
including life history traits (Falconer and Mackay
1996). While classical QTL mapping approaches
have been useful in determining the basic genetic
architecture of life history traits (e.g., the number
and effect size of the involved loci), they rarely
achieve sufficient resolution to pinpoint individual
candidate genes (see discussion in Roff 2007b and
Mackay et al. 2009). However, recent technological
advances, such as rapid and cost-effective genotyping methods and refined statistical and mapping
methods, have increased the feasibility of highresolution mapping, now allowing the identifica-
OUP CORRECTED PROOF – FINAL, 04/23/2011, SPi
I N T E G R A T I N G M E C H A N I S T I C A N D E V O L U T I O N A RY A N A LY S I S O F L I F E H I S T O RY V A R I A T I O N
tion of candidate genes within QTL regions for
organisms with well-annotated genomes, in some
cases down to the level of single nucleotide polymorphisms (e.g., Mackay et al. 2009). Highresolution mapping through recombinant inbred
lines and genome-wide association studies have
already been successful in characterizing natural
polymorphisms underlying genetic variation in
complex developmental or life history traits in C. elegans (e.g., Kammenga et al. 2007, Palopoli et al. 2008),
Drosophila (e.g., De Luca et al. 2003, Schmidt et al.
2008, also see Flatt and Schmidt 2009 for a recent
review), and in Arabidopsis (e.g., Atwell et al. 2010,
also see Chapter 9). Moreover, recent progress in
genomic methods now allows the researcher to treat
genome-wide expression patterns as complex quantitative traits (e.g., Rockman 2008).
The recent advent of refined QTL and genetical
genomics approaches is emblematic for an integrative and novel research program, namely the use of
natural genetic variation as a tool to understand the
causal connection between genotype and phenotype. By explicitly taking evolutionary variation
into account, this approach holds great promise for
facilitating the detection of mechanistic features
that are involved in phenotype construction.
However, the identification of individual genes or
nucleotide polymorphisms that contribute to quantitative trait variation remains a major challenge
because of subtle phenotypic effects, complex
genetic interactions, pleiotropy, and genotype-byenvironment interactions (e.g., Weigel and Nordborg
2005, Mackay et al. 2009).
1.4.4
How do trade-offs work?
One central and recurring theme in this book is the
mechanisms that underlie life history trade-offs (see
the chapters in Part 6). Given the central importance
of such trade-offs in life history evolution, uncovering their mechanistic basis is one of the most fundamental but unresolved problems in life history
research (e.g., Stearns 2000, also see Chapters 27
and 28). Despite numerous and seemingly obvious
trade-offs between life history traits in a wide range
of taxa, most reported trade-off relationships basically describe no more than a statistically inferred
negative correlation. The description of trade-offs
9
by means of trait correlations or covariances is,
however, insufficient for evaluating how genetic
architecture influences evolutionary trajectories
(e.g., see Chapter 2 and Roff 2007b). Specifically, it
remains to be determined to what extent presumptive trade-offs are conclusively due to actual competition for limited resources or caused by
alternative mechanisms, such as hormonal signaling independent of resource allocation (see Chapters
11, 13, 27, and 28). The very limited knowledge on
the mechanistic underpinnings of trade-offs therefore represents a current key problem in our understanding of life history evolution (e.g., Stearns 2000,
Flatt et al. 2005, Roff 2007b, Flatt and Schmidt 2009).
Recent progress in this area comes again from
the molecular genetic analysis of lifespan. Several
studies on the relationship between lifespan and
reproduction in worms and flies have challenged
the fundamental notion that reproduction exacts
an energetic cost in terms of reduced survival (e.g.,
see Chapter 11, Leroi 2001, Barnes and Partridge
2003). Of particular relevance was the observation
of a C. elegans insulin receptor mutant with extended
lifespan (Kenyon et al. 1993). Although this mutant
exhibited decreased fecundity—consistent with a
resource-allocation trade-off where investment in
longevity extension lowers investment in reproduction—detailed experimental analysis of this
relationship indicates that decreased reproduction
is not the causal agent in extending longevity (e.g.,
Kenyon et al. 1993, Leroi 2001). Therefore, reproductive versus somatic investment may not necessarily be coupled through resource competition but
rather via independent underlying signaling processes (see Chapters 11, 13, and 24, and Hsin and
Kenyon 1999, Flatt et al. 2008b). While these findings do not prove the absence of a cost of reproduction (Barnes and Partridge 2003, Flatt and Schmidt
2009), they underscore the difficulty of inferring
resource-allocation trade-offs without a precise
understanding of the proximate mechanisms
involved. For example, a major technical challenge
in demonstrating the resource basis of trade-offs is
to experimentally track resource allocation to different organismal functions by detailed measurement of relevant parameters, such as nutrient
ingestion and assimilation (see Chapter 24 and
O’Brien et al. 2008).
OUP CORRECTED PROOF – FINAL, 04/23/2011, SPi
10
M E C H A N I S M S O F L I F E H I S T O RY E V O L U T I O N
Other valuable information on the mechanistic
basis of life history trade-offs comes from research
exploring the fitness consequences of organismal
defensive mechanisms against pathogens, parasites,
stresses, or toxins. For example, studies in both
vertebrates and invertebrates indicate that elevated
immune and other defense functions incur fitness
costs in terms of reproduction and survival (see, for
example, Chapters 2 and 23, Flatt et al. 2005,
Harshman and Zera 2007). Similarly, the evolution
of pesticide tolerance in insects often results in a
fitness cost, which is generally supposed to stem
from increased energy allocation to corresponding
detoxification mechanisms. Remarkably, however,
it turns out that such fitness costs can result from
collateral metabolic costs rather than energetic costs
due to the detoxification mechanism (Van Straalen
and Hoffmann 2000).
Thus, while many observations support the existence and evolutionary relevance of life history
trade-offs, their underlying causal mechanisms still
remain rather poorly understood. Importantly, one
of the central postulates of life history theory,
namely that trade-offs are caused by competitive
resource allocation, might not necessarily always
hold. As discussed in many chapters throughout
this book (e.g., Chapters 11, 13, 27, and 28), major
efforts are currently under way to dissect the mechanistic basis of life history trade-offs.
1.5 Conclusions
Combining mechanistic and evolutionary analyses
of life history variation is a fundamental yet ambitious aim in current biology. On the one hand, there
are inherent biological and technical problems with
studying complex quantitative phenotypes such as
life history traits. On the other hand, there are
cultural divides that necessitate a combination of
diverse research approaches and concepts from
both molecular and organismal biology. Despite
these challenges, the chapters in this book illustrate
that the successful integration of mechanisms into
life history research is fully under way.